Fine-tuned w2v2 models for the Interspeech 2022 paper
Downloadable w2v2 models fine-tuned for SLU on MEDIA and FSC, used for the Interspeech 2022 paper.
See here for extracting features
See here for extracting features
Wav2Vec 2.0 models fine-tuned for SLU
| Model description | Link |
|---|---|
| W2V2-Fr-7K-large fine-tuned on MEDIA for SLU | Download |
| XLSR-53-56K fine-tuned on FSC for SLU | Download |